Nature Reviews Drug Discovery | AOP, published online 5 October 2015; doi:10.

1038/nrd4683

PERSPECTIVES
Specific criteria have been proposed for
OPINION
defining hits and leads in the development
of drugs for diseases such as malaria4–6;
Hit and lead criteria in drug however, it is essential that such criteria are
regularly reviewed and updated as part of
discovery for infectious diseases evolving target product profiles (TPPs) to
reflect accumulated experience from drug
of the developing world discovery projects and emerging scientific
research, clinical experience, policy guidance
and patient need. In order to reach a consen-
Kei Katsuno, Jeremy N. Burrows, Ken Duncan, Rob Hooft van Huijsduijnen, sus on the HTL criteria to be used to guide
Takushi Kaneko, Kiyoshi Kita, Charles E. Mowbray, Dennis Schmatz, its collaborative activities, the GHIT Fund
recently convened an initial gathering of its
Peter Warner and B. T. Slingsby
key international partners involved in drug
Abstract | Reducing the burden of infectious diseases that affect people in the discovery — the MMV, the TB Alliance and
developing world requires sustained collaborative drug discovery efforts. the DNDi — together with representatives
from the Bill & Melinda Gates Foundation.
The quality of the chemical starting points for such projects is a key factor in
The objective of this meeting was to take the
improving the likelihood of clinical success, and so it is important to set clear first steps towards creating a shared, flexible
go/no‑go criteria for the progression of hit and lead compounds. With this in strategy to expedite the discovery of the next
mind, the Japanese Global Health Innovative Technology (GHIT) Fund convened generation of drugs for these diseases by
with experts from the Medicines for Malaria Venture, the Drugs for Neglected GHIT Fund collaborations. Here, after briefly
discussing some of the general characteristics
Diseases initiative and the TB Alliance, together with representatives from the
of the relevant screening assays and estab-
Bill & Melinda Gates Foundation, to set disease-specific criteria for hits and leads lished target product profiles, we present the
for malaria, tuberculosis, visceral leishmaniasis and Chagas disease. Here, we proposed generic and disease-specific criteria
present the agreed criteria and discuss the underlying rationale. for hits and leads. We hope that these guide-
lines may also be valuable to drug discovery
There is an urgent need for new and chance of delivering a quality preclinical efforts outside these partnerships.
more-effective drugs to treat the various candidate4. The need to focus resources
diseases that take the heaviest toll on the effectively is particularly important in Hit discovery and assay development
developing world1. This can only be achieved drug research and development (R&D) For infectious diseases, hit candidates
in a cost-effective manner by implementing for infectious diseases that affect people in usually come from screens that involve intact
robust and efficient processes to develop the developing world, given the limited pathogens, akin to phenotypic screens (or,
and deliver drugs that are safe, effective, market incentives for R&D investment and more generally, high-content screens) rather
affordable and available to those who need the key role that philanthropic funding and than target-based screens. Which of these
them most. public–private partnerships have in this field. approaches is more productive is still being
The quality of chemical starting points The Japanese Global Health Innovative debated7,8. Despite being fuelled by advances
(known as ‘hits’) for drug discovery projects Technology (GHIT) Fund (BOX 1) is a pioneer- in genomics, target-based high-throughput
is a key factor for improving the likelihood ing public–private partnership. It provides screening, along with computer-assisted
of success of clinical candidates; starting a resources for research organizations in Japan modelling, has not been as productive as
discovery project with poor-quality hits ulti- to partner with international research groups phenotypic screening in the antibacterial area9.
mately results in increased attrition of these and product development partnerships However, for subsequent optimization efforts
compounds2,3. The decisions to progress a hit (PDPs) to screen and identify new hits and there is no doubt that knowing the molecular
into ‘lead’ identification (the HTL phase) and develop them into novel lead series for infec- target of a hit series is a major advantage. For
then on into lead optimization are crucial, tious diseases that affect people in the devel- instance, such knowledge allowed the clinical
as the downstream optimization phase may oping world. The initial areas of focus for the antimalarial candidate DSM265 to be specifi-
take years and require considerable financial GHIT Fund are malaria (in partnership with cally optimized to inhibit Plasmodium falci-
investment. Setting the bar high by applying the Medicines for Malaria Venture (MMV)), parum dihydroorotate dehydrogenase rather
comprehensive, well-considered criteria for tuberculosis (TB; in partnership with the than its human orthologue10. Additionally,
entry into lead optimization will improve TB Alliance), Chagas disease and visceral knowing the molecular target enables target-
overall success rates and focus resources leishmaniasis (in partnership with the Drugs specific liabilities to be identified earlier
on chemical series that stand a reasonable for Neglected Diseases initiative (DNDi)). in the R&D process. Furthermore, from a

NATURE REVIEWS | DRUG DISCOVERY ADVANCE ONLINE PUBLICATION | 1

© 2015 Macmillan Publishers Limited. All rights reserved

PERSPECTIVES

Box 1 | The Global Health Innovative Technology Fund: unlocking Japanese innovation
portfolio point of view, it ensures that a mix
of mechanistic approaches are followed,
The Global Health Innovative Technology (GHIT) Fund of Japan was founded in April 2013 with a which is especially relevant in diseases in
vision to alleviate the burden of infectious disease that prevents billions of people from seeking the which drug resistance develops and spreads
level of prosperity and longevity commonly expected in industrialized nations. Uniquely, the GHIT quickly (for example, TB and malaria).
Fund aims to do this by facilitating international partnerships that enable Japanese technology,
Fortunately, advances in genomics and
innovations and insights to have a more direct role in making this vision a reality67,68. Along with six
Japanese pharmaceutical companies and one diagnostic company, the GHIT Fund is supported by
related technologies usually enable the molec-
two Japanese government ministries and numerous unexpected co‑sponsors, including Mori ular target of a drug (or at least its mode of
Building Co., Ltd., All Nippon Airways Co., Ltd., Morrison & Foerster LLP, and Yahoo! Japan (see the action) to be elucidated even when the com-
figure, part a). The Japanese pharmaceutical industry is represented by the 69 members of the pound was discovered in a phenotypic screen;
Japanese Pharmaceutical Manufacturers Association. such screens may therefore be viewed as
A key aim of the GHIT Fund is to link existing product development partnerships (PDPs) — discovery engines for druggable targets, along
responsible for discovering and delivering new medicines for malaria, tuberculosis (TB) and with the drug molecules themselves. The two
neglected tropical diseases (NTDs) — with Japanese partners that have expertise in drug research crucial success components for screening
and development (R&D). This is achieved by establishing a portfolio of screening projects and a are the availability of chemical libraries with
hit‑to‑lead (HTL) platform that use defined criteria to ensure that the most attractive chemical series
‘interesting’ chemistry and taking great care
are selected for further optimization and development. Research collaborations must comprise at
least two distinct organizations, one of which is Japanese, in order to be eligible. For the screening
to develop assays that faithfully reproduce the
and HTL platforms the compounds must originate from a Japanese entity. In addition, the HTL microenvironment of a pathogen with a dis-
platform aims to maximize the skills and resources of existing global PDPs that operate in the field of ease-relevant readout, while also maintaining
NTDs and other infectious diseases that affect people in the developing world, such that each new throughput and robustness. Such advanced
partner must enter into a research collaboration with one of three specific PDPs (the Medicines for assays also eliminate compounds that do not
Malaria Venture (MMV), the Drugs for Neglected Diseases initiative (DNDi) or the TB Alliance). penetrate cellular membranes or are otherwise
Beyond the benefits of sharing expertise, collaborations of this type will allow diverse chemical unavailable to the target.
libraries to be probed using various screening approaches. Importantly, Japanese compound assets The organisms that cause malaria,
of synthetic and natural origin could be a rich source of novel and chemically diverse compounds TB and NTDs are biologically quite diverse,
that have not previously been screened for infectious diseases of the developing world. Past
including viruses, bacteria and eukaryotes.
examples of key natural-product-based drugs that originated from Japanese discovery efforts
include ivermectin, an antiparasitic drug from the actinomycete Streptomyces avermectinius69–71;
Each of these comes with its own challenges
the multiple sclerosis therapy fingolimod, which is a metabolite of the insect fungus Isaria sinclairii72; and opportunities, resulting in differences
and the founder of the statin drug class, mevastatin, which was discovered in Penicillium spp. in hit and lead criteria and assay conditions.
extrolites by Akira Endo of Sankyo in the 1970s73,74. Dogma holds that bacteria replicate quickly
In just over 2 years since its inception, the GHIT Fund has facilitated more than 30 partnerships (~20‑minute generation time) and eukaryotic
(for screening collaborations see the figure, part b) and invested more than US$40 million in them. (mammalian) cells double in number in
All proposals received are evaluated by external reviewers and a selection committee, both of which ~24 hours, but these guidelines do not apply
are completely independent from the GHIT Fund. to the pathogens that cause the four diseases
In addition to the existing three platforms (the screening, HTL and product development platforms), discussed in this article. Mycobacterium
the GHIT Fund also initiated the ‘Target Research Platform in Partnership with Grand Challenges’
tuberculosis is an exception among bacteria
(TRP) for early-stage development of radically new and improved drugs, vaccines and diagnostics
to prevent and treat infectious diseases that are prevalent in developing countries. With additional
in that it has up to 16‑hour doubling times,
Japanese partnerships expected and additional compound screening and HTL proposals being whereas the doubling times of the eukaryotic
submitted67,68, the GHIT Fund encourages research and shows a commitment to the cause that is Leishmania spp. parasites, which cause
unprecedented in this area of research within Asia. visceral leishmaniasis, are ~6–9 hours.
a The in vitro culture of all major strains of
Public sector Private sector Civil sector Co-sponsors P. falciparum — the species responsible for
• Japanese Ministry • Astellas Pharma, Inc. • Bill & Melinda • Mori Building the majority of deaths from malaria — has
of Foreign Affairs • Chugai Pharmaceutical Gates Foundation Co., Ltd been an important breakthrough11, but its
• Japanese Ministry Co., Ltd • United Nations • All Nippon doubling time is ~48 hours12. Owing to these
of Health, Labour • Daiichi Sankyo Development Programme Airways Co., Ltd.
and Welfare Company, Limited
differences, biomass generation and assay
• Wellcome Trust • Morrison &
• Eisai Co., Ltd. Foerster LLP set‑up vary in difficulty between organisms.
• Shionogi & Co., Ltd. • Yahoo! Japan The potency cut-offs for compounds to
• Takeda Pharmaceutical progress to the next stage of development
Company Limited varies by disease (see below) and is decided
• Sysmex Corporation by a number of factors. Among these
factors are empirical hit rates (the higher the
o

hit rate, the lower the cutoff concentration).

M are nky

Plasmodium spp. hit rates are generally good,

v a

ip o
i
be ish

b
No hi-S
F

in tom
n
yo -

CR
nk hi

po
ct io

na b
gi

which is possibly related to the unusual (and

to

Ta itsu
y
RD iic
Sa aiic

Fa P B

Da umi
M

no

or
sa

ed
C–

Da
i

io

vital) apicoplast organelles that these organ-

ta

O
D

sa

S
k
IM

Sh
Ki

Ta
Ei

DNDi isms harbour. Moreover, these organisms

MMV must actively overcome the toxicity of haem
TB Alliance degradation and remodel their host cell (red
blood cell) membrane transport capabilities,
IMC, Institute of Microbial Chemistry; MCRF, Microbial Chemistry Research Foundation. and these processes are druggable. By contrast,
Nature Reviews | Drug Discovery

2 | ADVANCE ONLINE PUBLICATION www.nature.com/reviews/drugdisc

© 2015 Macmillan Publishers Limited. All rights reserved


it is perhaps not surprising that targeting
and killing the intracellular amastigote form
of Leishmania spp. parasites is more chal-
lenging, as amastigotes reside inside acidic
phagolysosomes within macrophages, which
present additional membranes and pH gra-
dients that drugs must cross before reaching
their target. Trypanosoma cruzi, the parasite
that causes Chagas disease, presents its own
challenges owing to its ability to infect many
cell types, producing a highly dynamic infec-
tion. Furthermore, drugs that only target
the replicating stages of the parasite may
leave non-replicating forms, such as trypo­
mastigotes, capable of maintaining infections
long after the end of the treatment13.
Established target product profiles
It is important that hit and lead series are
assessed as early as possible for conformity
with the relevant disease TPPs and, if appro-
priate, target candidate profiles (TCPs).
However, in most cases, further detailed
biological or pharmacokinetic studies will
be required to fully judge how a series can be
strategically positioned. These TPPs and
TCPs are typically developed by the PDPs
in discussion with the research and medical
communities.
TPPs and TCPs for malaria. In the case of
malaria, two TPPs and five TCPs have been
established, which reflect different patient
populations and medicinal uses6,14. In brief,
malaria TPP1 is focused on treatment, ideally
from a single dose, such that all symptomatic
and asymptomatic parasites (gametocytes) in
a host are cleared and transmission is blocked
in addition to the patient being cured. TPP2
is focused on prophylaxis, because when
attempting to eliminate a disease it is recog-
nized that people in once-endemic regions,
now with potentially reduced immunity, may
require protection in the event of transmis-
sion outbursts. The TPPs focus on the profile
of a medicine that is composed of two or
more active ingredients (a criterion that is
presently mandatory for artemisinins15).
By contrast, the TCPs focus on attributes
that individual molecules need to possess
(acknowledging that one molecule may
possess more than one attribute). The key
features of the five TCPs are: fast parasite
clearance (TCP1); a combination partner,
ideally long duration, which provides post-
treatment prophylaxis (TCP2); prevention
of Plasmodium vivax and Plasmodium ovale
relapse (TCP3a); prevention of transmission
(TCP3b); and chemoprotection (TCP4)6.
TCP1, TCP2, TCP3a and TCP3b all support
TPP1, whereas TCP4 supports TPP2 (REF. 6).
NATURE REVIEWS | DRUG DISCOVERY ADVANCE ONLINE PUBLICATION | 3
PERSPECTIVES
Box 2 | Generic hit selection criteria for infectious diseases
The panel of drug experts convened by the Japanese Global Health Innovative Technology (GHIT)
Fund identified the following generic hit selection criteria:
• A hit should have a potency that is consistent with the potential to deliver a lead compound
(see BOXES 4–6 for details of criteria defined specifically for each disease)
• The chemical structure of a hit should be confirmed by identification (for natural products),
re‑synthesis or re‑purification
• Primary screening data should be validated on a selection of hit compounds (>90% pure)
• A hit should have an acceptable in vitro response (typically, a sigmoidal concentration–growth
inhibition curve reaching a maximal 100% efficacy, with a Hill coefficient ideally between 0.5
and 1.8)
• Preliminary knowledge of the structure–activity relationship (SAR; often available from
analogues in the original screening library) of a hit is desirable
• A hit should have a tractable chemotype: it should have no highly reactive or unstable moieties
in the pharmacophore and be amenable to structural variation by chemical (or biochemical)
synthesis. Hits should pass basic drug-like filters, such as pan-assay interference filters (PAINS)75,
to eliminate promiscuous hits that lack target specificity76,77. Conformity to the ‘rule of five’
(REF. 58) is preferred
• There should be a greater than 10‑fold selectivity window for cytotoxicity using a mammalian
cell line (for example, HepG2 or Vero cells)
• A hit requires adequate selectivity in a biochemical counter-assay (for example, a homologous
mammalian target) where relevant. However, most infectious disease hit-to-lead projects are
not target-based screens but phenotypic
• No serious intellectual property conflicts should exist (that is, a ‘freedom to operate’ is needed).
However, with the value of US Food and Drug Administration (FDA) priority review vouchers now
entering the pharmacoeconomic equation, there are further possibilities to develop drugs for
infectious diseases even in absence of intellectual property protection
• No major synthesis or formulation issues should be anticipated (compounds should ideally
be synthesized in ≤5 steps with an acceptable yield and acceptable solubility). For reasons of
affordability, this criterion is more stringent than for drug discovery in general
TPPs for TB. TPPs for TB usually require Generic criteria for hits and leads
new drugs to shorten the duration of treat- Hit series definition and criteria. A number
ment, demonstrate efficacy against drug- of generic hit criteria, identified by a panel
sensitive and drug-resistant M. tuberculosis of drug experts convened by the GHIT
strains and show potential for use in drug Fund, apply to all four infectious diseases,
combinations in developing countries and these are listed in BOX 2. In general,
(for example, those with oral and once-daily the objectives at this stage are to build
dosing, and low cost of goods). These goals confidence in the quality of a compound
are based on the fact that the majority of TB series and the associated data, understand
drugs and treatments have not changed in the liabilities and, if possible, generate data
the past half-century, even though two new that guide medicinal chemists during the
drugs (bedaquiline16 and delamanid17) were HTL phase18. Two main types of criteria
recently approved — the first new drugs for are covered: disease-specific criteria, focus-
TB in the past 40 years. Additional drugs are ing on potency, efficacy and pathogenicity;
needed to develop new treatment regimens. and compound-specific criteria, which
focus principally on the chemical scope
TPPs for Chagas disease and leishmaniasis. of the compound and a risk assessment of
The DNDi has defined a TPP for Chagas drug metabolism and pharmacokinetics
disease with ‘acceptable’ and ‘ideal’ criteria (see (DMPK), as well as the physical properties
the DNDi website). The benchmark for clini- that are predictive of an oral therapy.
cal efficacy is benznidazole, and this should
therefore be the comparator for all hit and lead Lead series definition and criteria. During
candidates. Similarly, the DNDi has published the HTL phase, the project team focuses on
(with consultancy) ‘optimal’ and ‘minimal’ the optimization of a chemical series so as
TPPs for visceral and cutaneous leishmaniasis. to improve any compound properties that
For visceral leishmaniasis, a safe oral drug could be an obstacle to further progression.
with >90% efficacy within 10 days is crucial. Depending on the profile, the HTL strategy
Relevant for early compound triage is that a and medicinal chemistry plan can be very
drug must be active against all resistant strains. different between series. The milestone for
© 2015 Macmillan Publishers Limited. All rights reserved
PERSPECTIVES

Box 3 | Generic lead selection criteria for infectious diseases

Breaking the life cycle and killing parasites
during other asymptomatic phases — such as
The panel of drug experts convened by the Japanese Global Health Innovative Technology (GHIT) the sexual-stage gametocytes and liver-stage
Fund identified the following generic lead selection criteria: hypnozoites and schizonts — will be crucial
• A lead requires an acceptable in vitro potency for the relevant disease (see BOXES 4–6 for for providing treatments to block relapse
disease-specific details). In general, high potency (a low half-maximal inhibitory concentration in P. vivax infections, preventing transmis-
(IC50)) is highly desirable but not at the expense of poor physicochemical properties or drug
sion and protecting vulnerable populations.
metabolism and pharmacokinetic (DMPK) characteristics
Indeed, only with such drugs can the global
• A lead should have oral efficacy in the appropriate disease model. Oral efficacy removes one key
risk early on, namely, uncertainty about in vivo validation of the mechanism of action of the series
goal of eradicating malaria be realized.
(which is often unknown at this stage) Extensive research on these two disease-causing
• The synthetic chemistry should be amenable to series expansion, as many more compounds are pathogens (P. falciparum and P. vivax) over
likely to be needed for testing in as short a time frame as possible the past decade has helped to establish some
• A lead needs a greater than 10‑fold selectivity in killing pathogens as opposed to mammalian cells very specific entry criteria for drug discovery
in tuberculosis (TB) and a greater than 100‑fold selectivity for the pathogens that cause the other programmes14.
diseases (malaria, Chagas disease and visceral leishmaniasis). This reflects the increased challenge The cellular potency criterion for a hit
of finding potent, attractive compounds that target TB for malaria research (BOX 4) is based on the
• A lead should have acceptable physicochemical properties (typically, solubility in extensive screening efforts of the MMV and
phosphate-buffered saline >10 μM; a sufficient level of solubility is expected to avoid problematic its partners over the past 7 years24. Through
formulations78. Acceptable lipophilicity; LogP values are typically <5 and ideally <3)58 partnerships in both industry and academia,
• A lead requires manageable drug metabolism and pharmacokinetic profiles. This involves liver more than 5 million compounds have been
microsomal and hepatocyte stability across species; understanding of the plasma protein, screened against the asexual blood stages of
microsomal and media binding; good membrane permeability; and no unmanageable cytochrome P. falciparum and the cutoff potency from
P450 inhibition
these in vitro screens has mostly been around
• A lead needs to demonstrate good oral bioavailability in rodents (demonstrated F>25%); 1–2 μM. Nevertheless, more than 25,000 com-
ultimately, high oral bioavailability is highly desirable as it will reduce the potential for inter-patient
pound hits were available for follow-up25–27.
variability, pill size, dose and cost of the medicine. It is therefore crucial that this parameter is
tractable at the lead stage Screening against liver stages of Plasmodium
• A lead needs an acceptable early safety assessment based on target (orthologue) and compound
spp. has also delivered many potent hits28.
liabilities, in vivo observations, in vitro studies (for example, genotoxicity and the mini-Ames test), Naturally, activity should be confirmed with
cytotoxicity, cardiac safety (as assessed using the hERG channel (QT prolongation)) and in silico a pure compound, and hits should have selec-
approaches. A secondary pharmacology selectivity profile, consistent with achieving selectivity tivity for the parasite over a mammalian cell
with the candidate compound, is also required. This would include human orthologues and line (for example, a greater than 10‑fold
paralogues of the targeted enzyme or receptor, if known; the number of these that are to be tested difference between the IC50 and CC50, the
depends on the gene family size and their known or suspected safety risks (when targeted) half-maximal inhibitory concentration against
• All liabilities of the series should be understood (as a result of extensive profiling) and a rationale the parasites and the half-maximal cytotoxic
and medicinal chemistry plan generated for how they might be overcome in the subsequent concentration against the host mammalian
optimization phase cells, respectively) and display an acceptable
• A lead should contain no known toxicophores or undesirable reactive groups and no detrimental concentration response, all of which provide
chemical feature or characteristic associated with the pharmacophore indicative of, for example,
confidence in a specific interaction and effect.
adduct formation. This avoids the scenario in which a region of the molecule responsible for
By the start of lead optimization there
activity has a liability that cannot be overcome
is a need for clarity on the TPPs and TCPs of
• A lead should display no acute toxicity in in vivo efficacy studies. Although no formal in vivo safety
studies are performed at this stage, careful observation of efficacy studies, particularly at high,
the series, as these define the goal and tactics
repeated doses, can be informative for the subsequent optimization phase.
• Preferably, there should be no apparent intellectual property obstacles to the progression The potency required for progression
of any series (freedom to operate) depends on the TCP. For blood stages (TCP1
and TCP2), the crucial aspect is in vivo oral
efficacy in the P. falciparum SCID (severe
this phase is the delivery of a lead series. in South-East Asia where cases of increased combined immunodeficiency) mouse
The generic criteria for such a lead are parasite-clearance times are on the rise in model of infection29, with the key feature
shown in BOX 3 (REF. 19). patients treated with combination therapies for TCP1 compounds being rapid parasite
that include artemisinin derivatives (for clearance in vivo, at rates at least as good
Disease-specific hit and lead criteria example, artesunate). This has revealed a as those of chloroquine. For prophylaxis
Guided by the specific requirements for each new type of Plasmodium spp. resistance — (TCP4), the crucial aspect is in vivo oral
disease as well as existing target product and essentially one in which the ring stage of the efficacy in a Plasmodium berghei sporozoite
candidate profiles, the committee coordinated intra-erythrocytic cycle can tolerate drug challenge model (or equivalent)30. From
by the GHIT Fund devised disease-specific intervention20–23. As the mainstay malaria experience, achieving a dose that eradicates
criteria for hits and leads, which are discussed treatments rely on a component that is 90% of the target pathogen (ED90) <50 mg
below and summarized in BOXES 4–6. artemisinin-based, a global public health per kg provides confidence that the series
catastrophe could emerge unless new has a parasitological foundation for the lead
Malaria. There are several good treatments drugs that overcome this risk are delivered. optimization phase; ultimately, a success-
for malaria, but the challenge of emerging In addition, current antimalarials generally ful candidate will have an ED90 <10 mg per
drug resistance is ever present, particularly target the asexual blood stage of the disease. kg (often considerably lower). Typically,

4 | ADVANCE ONLINE PUBLICATION www.nature.com/reviews/drugdisc

© 2015 Macmillan Publishers Limited. All rights reserved

 PERSPECTIVES

this is achievable if a compound has an IC50
<100 nM against the blood or liver stages
and appropriate DMPK properties. However,
a compound may still be acceptable for
follow up if one of these components does
not meet the proposed criteria, provided
the other parameter is outstanding.
Given the risk of resistance and a need for
drugs with novel modes of action, potency
in vitro across a panel of established drug-
resistant parasite strains isolated from patients
is also crucial. The genetics underlying the
emergent resistance against artemisinins, as
seen in the Mekong area in South-East Asia,
are becoming increasingly understood31–33,
and this knowledge is being used to set up
panels of parasites against which new drug
candidates can be tested. For TCP1 at least,
new drugs are most likely to be used in
combinations (as presently mandated for all
artemisinins). This means that a newcomer
must be evaluated for potential drug–drug
interactions with other TCP1 antimalarials.
For TCP3a (anti-relapse), it is necessary
to have in vitro data supporting the activity
of compounds on hypnozoites34, whereas for
TCP3b additional activity against mature and,
ideally, early-stage gametocytes is required in
addition to asexual blood-stage potency 35.
Box 4 | Summary of main checkpoint criteria for antimalarial hits and leads
Guided by the specific requirements for the disease, and taking into account existing target product
and candidate profiles, the committee coordinated by the Global Health Innovative Technology
(GHIT) Fund devised the following disease-specific criteria for hits and leads for malaria.
Validated hit
• Cellular potency criteria: hits should have an effector concentration for half-maximum
response (EC50) <1 μM for sensitive and multiple resistant strains of Plasmodium spp.
• Cytotoxicity criteria: hits require a greater than 10‑fold selectivity between the half-maximal
cytotoxic concentration (CC50) for the mammalian cell line and the EC50 for Plasmodium spp.
(See also BOX 2 for the evaluation criteria regarding the acceptability of chemical structure,
novelty and confirmation)
Early lead
• Cellular potency criteria: a lead requires EC50 <100 nM for sensitive and multidrug-resistant
strains of Plasmodium spp.
• Cytotoxicity criteria: a lead should have a greater than 100‑fold selectivity between
mammalian cell line CC50 and Plasmodium EC50. Frontrunners should be tested across the
malaria life cycle and key mechanistic assays so as to ensure an understanding of the
phenotype and target candidate profile (TCP) potential of each series and (preferably) novel
mechanisms of action
• In vivo efficacy criteria: when administered orally in the blood stages of infection (TCP1 and
TCP2), a lead should achieve parasite clearance at a dose that eradicates 90% of the target
pathogen (ED90) <50 mg per kg (typically four doses over 4 days) in the Plasmodium falciparum-
infected SCID (severe combined immunodeficiency) mouse model. TCP1 should demonstrate a
rapid rate of parasite clearance. For TCP3a, the anti-relapse TCP, there are no in vivo criteria for
leads (in the absence of good models), but a lead should demonstrate anti-hypnozoite activity
in vitro. For TCP3b, the transmission-blocking TCP, a lead should demonstrate potency in a
gametocyte assay (for example, as measured by gamete formation79) in a similar range to that
of the in vitro asexual blood stage potency. For TCP4, the chemoprotection TCP, a lead should
have efficacy in a prophylaxis model of malaria, with ED90<50 mg per kg

Tuberculosis. M. tuberculosis is so well

adapted to its human host that almost one
third of the world population is estimated to
be infected. Although only 10% of infected
individuals are believed to develop TB
in their lifetime, this still results in ~1.5
million deaths annually 36 (see the World
Health Organization (WHO) website).
Therefore, there is an urgent need for new
anti‑TB drugs.
It still takes 6 months to cure drug-sensitive
TB (DS-TB) and a minimum of 18 months
to treat multidrug-resistant TB (MDR-TB).
It is highly desirable to shorten treatment
duration for both patients with DS-TB and
those with MDR-TB to improve compliance
and to limit the spread of drug-resistant TB
(DR-TB). Standard care of patients with
DS-TB includes a combination of isoniazid,
rifampicin, pyrazinamide and ethambutol
for the first 2 months and a combination of
isoniazid and rifampicin for the remaining
4 months. Patients with MDR-TB, whose
M. tuberculosis strains are resistant to isonia-
zid and rifampicin, are treated with second-
line TB drugs that include aminoglycosides,
quinolone antibiotics, cycloserine and
capreomycin. An updated dataset was
recently published for the most commonly
used TB drugs with respect to in vitro
potency, cidality, physicochemical and
pharmacokinetic properties generated
under standardized conditions9. For patients
treated for MDR-TB, the cure rate is only
~48% and this needs to be improved
drastically, whereas the cure rate of
patients treated for DS-TB is ~85%9.
There has been a steady increase in TB
drug resistance; the WHO estimated that
3.5% of new cases and 21% of previously
treated cases are MDR-TB36. Among patients
with MDR-TB, an estimated 9% have exten-
sively drug-resistant TB (XDR-TB), meaning
that the pathogens are resistant to all second-
line TB drugs36 in addition to rifampicin and
isoniazid. A new drug must show efficacy
against all resistant strains, which calls for
agents with novel mechanisms.
The new candidates must also be orally
available to ensure wide usage, especially in
developing countries. For the same reason,
the cost of goods needs to be low and pill
size (which is determined by potency and
pharmacokinetic properties) reasonable.
Furthermore, as they will be used in com-
bination with other TB drugs to stem the
emergence of resistance, the new drugs must
have a low risk for drug–drug interactions.
As patients with TB are often co‑infected
with HIV, new agents also need to be com-
patible with most of the anti-retroviral
drugs. All of these requirements are
especially stringent because TB drugs are
administered for extended time spans.
Treatment shortening can only be achieved
with an agent or a regimen that effectively
eliminates non-replicating M. tuberculosis9.
Non-replicating M. tuberculosis is metaboli-
cally less active and less susceptible to anti­
biotics than actively replicating bacteria9,
and new hits must be evaluated for their
capacity to kill both replicating and non-
replicating M. tuberculosis, as measured in
the microplate Alamar Blue assay (MABA37)
and low oxygen recovery assay (LORA38),
which are both suited for high-throughput
screening. The desired hit and lead profiles
are summarized in BOX 5.
The most-recent TB hits were discovered
in phenotypic screens with cultured M. tuber-
culosis, typically using the MABA9. In addi-
tion, low-oxygen culture conditions (LORA38)
and other assays39,40 are used to identify hits
that kill slow- or non-replicating bacteria41.
However, these assays do not fully capture the
little-understood mechanisms whereby TB
bacteria move in and out of latency 42–44.
In addressing this need, a new assay was
recently developed for agents that kill M. tuber-
culosis bacteria that reside in macrophages45,
resulting in the discovery of Q208 (REF. 46).

NATURE REVIEWS | DRUG DISCOVERY ADVANCE ONLINE PUBLICATION | 5

© 2015 Macmillan Publishers Limited. All rights reserved

PERSPECTIVES

Box 5 | Main checkpoint criteria for anti-tuberculosis hits and leads

resistance. Furthermore, a treatment adapted
to field conditions, with a shorter treatment
Guided by the specific requirements for the disease, and taking into account existing target product duration and that could be used globally,
and candidate profiles, the committee coordinated by the Global Health Innovative Technology would be optimal. In addition, to address
(GHIT) Fund devised the following disease-specific criteria for hits and leads for tuberculosis (TB). the development of resistance by the parasite
Validated hit to drug monotherapies — as documented
• Cellular potency criteria: a hit should have a minimum inhibitory concentration (MIC) against for current therapies — drugs with new and
the Mycobacterium tuberculosis laboratory strain H37Rv of <10 μM under replicating growth distinct mechanisms of action should be
conditions, for example, under microplate Alamar Blue assay (MABA) conditions37 developed as combination therapies early in
• A hit requires a greater than a 10‑fold difference between cytotoxicity against mammalian cells development and not used as monotherapies.
(Vero cells) and the MIC against M. tuberculosis H37Rv In the case of Chagas disease, there are
• Hits should show evidence of a preliminary structure–activity relationship (SAR) among analogues even fewer treatment options than there are
(See also BOX 2 for the evaluation criteria regarding the acceptability of chemical structure, novelty for visceral leishmaniasis. Monotherapy with
and confirmation) nifurtimox or benznidazole (both from
Early lead the same nitroheterocycle class) remains the
• In vitro potency criteria: a lead should have a MIC against the M. tuberculosis strain H37Rv of only recognized treatment, but these agents
<1 μM under MABA conditions and a MIC <10 μM under non-replicating anaerobic conditions require long treatment courses, have variable
(low-oxygen-recovery assay, LORA38 conditions) efficacy and cause serious side effects,
• A lead should exhibit in vitro activity against M. tuberculosis strains that are resistant to a single resulting in discontinuation of treatment in
TB drug, such as isoniazid or rifampicin, indicating a new mechanism of action ~20–30% of patients. It is crucial that new
• A demonstration of the in vitro bactericidal activity of a lead is required as indicated by classes of effective, well-tolerated, orally acting
time­–kill curves showing that the number of colony-forming units (CFUs) decreases over time and short-course treatments are progressed
(for example, 14 days) compared with the number of CFUs at the beginning of the experiment into the clinic to provide improved options
• The oral bioavailability of a lead must be demonstrated in rodents for patients. New classes of drugs for Chagas
• A demonstration of the oral efficacy of a lead in a mouse acute infection model is required80 disease are also essential to enable the devel-
• A preliminary indication of the safety of a lead must be demonstrated (in hERG81 and cell opment of combination therapies to improve
health assays82) efficacy and toleration, reduce treatment dura-
tion and combat the risk of the development
of resistance to monotherapies. Recently, a
The second challenge in discovering drugs The goal for research must be to transform trial in chronic Chagas disease was described
that target latent TB requires an understanding patient therapy from poorly adapted anti­ for posaconazole, an inhibitor of T. cruzi 14‑α
of TB pathogenesis. M. tuberculosis primarily monial treatments (for example, sodium demethylase (CYP51) that has a different
infects and replicates in activated macro­ stibogluconate (SSG)) to simple, patient- mechanism of action to the currently used
phages but can persist in a non-replicating adapted oral therapies that are affordable, safe nitroheterocyclics, as exemplified by benzni-
state in foamy macrophages47. This host– and efficacious in both children and adults. dazole53. In spite of good preclinical efficacy,
pathogen system can form granulomas in In the past 15 years, combinations of the new compound did not meet expectations,
which infected macrophages are surrounded drugs with similar or improved efficacy to resulting in more treatment failures than
by layers of active macrophages and other the older antimonials, but with improved benznidazole. It is believed that the standard
immune cells. The central granuloma region is safety and tolerability profiles, have been Chagas mouse model that was used to validate
oxygen-deprived and necrotized but contains developed. These include combinations posaconazole only represents the early acute
latent M. tuberculosis; in the lung this results containing liposomal amphotericin B, paro- phase of infection; thus, it is important to use
in cavities. The inner necrotic area is poorly momycin and/or miltefosine52. However, models and assays that also allow testing of
irrigated, complicating local drug exposure. these drugs remain costly, are difficult to the chronic late stage of the disease13,54.
Granulomas do not normally form in in vivo administer, have poor stability at the high In order to promote the discovery of safe,
models; however, recently, animal models temperatures that occur in endemic regions, efficacious and orally acting treatments for
of granulomas were developed48,49. In one of require lengthy treatment and/or are poorly visceral leishmaniasis and Chagas disease
these models, it was shown that treatment tolerated. In addition, there is a dichotomy that overcome the limitations of existing
with a vascular endothelial growth factor in drug efficacy in regions of the world regimens, discovery needs to focus on new
(VEGF)-specific antibody can restore vascu- where visceral leishmaniasis is endemic. chemical series, leaving behind the flawed
lature and drug access in TB granulomas49. In South Asia, the medical needs for visceral classes in current use. Until recently, the
Continued research into new assays and mod- leishmaniasis are presently moderately to steady identification of new chemical series
els is crucial, as well as pursuing completely well met. However, in East Africa and Latin has been hindered by a limited screening
different approaches that aim to stimulate host America, the efficacy and tolerability of capacity coupled with very low hit rates.
immune responses against TB50. current visceral leishmaniasis therapies However, progress has been made and
remain a challenging area for improvement. larger compound collections are now being
Visceral leishmaniasis and Chagas disease. Ideally, what is needed in the treatment screened. More than a million compounds
For visceral leishmaniasis, existing drugs have of visceral leishmaniasis is a simple oral have been screened against T. cruzi and
variable efficacy and serious toxicities51; only combination therapy that would prove to be Leishmania spp.54–56, but the rate of hit dis-
one (miltefosine) is administered orally and advantageous and/or effective in maintaining covery still lags far behind that for malaria.
the others are given by intravenous or intra- or improving efficacy, improving tolerability It remains necessary to avoid discarding
muscular injections, which are impractical. and preventing or delaying the emergence of precious chemical series by setting the hit

6 | ADVANCE ONLINE PUBLICATION www.nature.com/reviews/drugdisc

© 2015 Macmillan Publishers Limited. All rights reserved


selection criteria too high. Thus, hit criteria
focus on the identification of new series
and mechanisms of action with even
modest activity against the intracellular
forms of Leishmania donovani and T. cruzi,
the causative agents of leishmaniasis and
Chagas disease, respectively 57.
Although this inclusive approach to hit
selection provides scope for subsequent
HTL projects it does bring with it a high
rate of early attrition and often leads to long
optimization campaigns before quality leads
are produced. The HTL entry criteria stated
in BOX 6 are a combination of the current
well-documented ‘best practices’ for targeted
drug discovery projects58 and a modest
phenotypic in vitro activity hurdle59. Once
the encouraging in vitro anti-parasitic activity
of a new series can be coupled with sufficient
in vivo plasma exposure, ideally following
oral dosing, the next hurdle is to demonstrate
a robust reduction in the parasite burden at
target organs in infected rodents, which is the
defining characteristic of a lead that is ready
for subsequent optimization. Unfortunately,
we have little knowledge regarding how
in vitro potency correlates with in vivo activity
in visceral leishmaniasis and Chagas
disease, with a lack of pharmacokinetic and
pharmacodynamic (PK/PD) examples. For
T. cruzi infection, oral dosing is required for
in vivo proof‑of‑concept studies, which is in
line with the requirement for orally acting
therapies. Although oral dosing is also pre-
ferred for studies with rodents infected with
Leishmania spp., intraperitoneal or intra-
venous administrations are also acceptable
given that a short-course treatment adminis-
tered by the parenteral route could fulfil the
TPP for visceral leishmaniasis.
Discussion
Our consultation among drug discovery
experts collaborating with the GHIT Fund
revealed differences in the criteria consid-
ered most conducive to cost-efficient drug
discovery, which are associated with the spe-
cific requirements for the diseases reviewed
here. One of the dangers in establishing
stringent, clear-cut criteria for entry into the
HTL phase and then progression into lead
optimization is that valuable chemicals are
discarded at an early stage. It is obvious that
a molecule such as artemisinin, were it to
emerge today as a hit in a high-throughput
screening campaign, would severely struggle
to be taken forward, scoring extremely low
in chemical tractability, ease of synthesis
and chemical suitability (the compound
has a highly reactive peroxide). Even if it
were taken further (without analysis of the
NATURE REVIEWS | DRUG DISCOVERY ADVANCE ONLINE PUBLICATION | 7
PERSPECTIVES
Box 6 | Hit and lead criteria for Chagas disease and visceral leishmaniasis
Guided by the specific requirements for the disease, and taking into account existing target
product and candidate profiles, the committee coordinated by the Global Health Innovative
Technology (GHIT) Fund devised the following disease-specific criteria for hits and leads for
Chagas disease and visceral leishmaniasis.
Hit and lead criteria for Chagas disease
• Key hit selection criteria: cellular potency — a hit should have a half-maximal inhibitory
concentration (IC50) <10 μM against intracellular Trypanosoma cruzi
• Key lead selection criteria: in an acute mouse model of Chagas disease, a hit should demonstrate an
80% reduction of parasite burden in organs or tissues, or there should be no parasites detected at
the end of treatment and an increase in lifespan with up to 10 doses at 50 mg per kg delivered orally
Hit and lead criteria for visceral leishmaniasis
• Key hit selection criteria: cellular potency — a lead should demonstrate an IC50 <10 μM against
intracellular Leishmania donovani
• Key lead selection criteria: in a mouse (or hamster) model (infected with L. donovani or
Leishmania infantum), treatment with a lead should result in a >70% reduction in liver parasite
burden after at most 5 doses at 50 mg per kg delivered orally once or twice per day
structure–activity relationship (SAR)), considerations presented here in the light of
scientists would discover that the compound the four sets of disease TPPs as beacons.
has a very short in vivo half-life (minutes), is Our analysis has split recommendations for
metabolically unstable and its mode of action entry into HTL and lead optimization phase
and target are poorly understood60. Along into a generic part and a more specific part
with serious preclinical safety concerns61,62, for each disease. As stated earlier, these guide-
there is every reason to believe that the lines were established in the context of GHIT
modern drug discovery process would have Fund-coordinated collaborations, but we
discarded the drug long before it had a chance hope that they will find wider adoption and
to save the lives of millions. In fact, algorithms aid the acceleration and expansion of pipeline
that select ‘drug-like’ molecules generally drug candidates for these serious diseases.
exclude artemisinin-like molecules from the
Kei Katsuno and B. T. Slingsby are at the Global Health
chemical libraries used for HTS in the first Innovative Technology (GHIT) Fund, Ark Hills,
place. Chemical tractability is important, but Sengokuyama Mori Tower (25F), 1‑9‑10 Roppongi,
it must be kept in mind that there are many Minato‑ku, Tokyo 106–0032, Japan.
other molecules that have progressed straight Jeremy N. Burrows, Rob Hooft van Huijsduijnen and
from a phenotypic screen (that is, without Dennis Schmatz are at the Medicines for
further chemical modification) into patients, Malaria Venture (MMV), 20, Route de Pré-Bois,
1215 Geneva 15, Switzerland
such as tamatinib (also known as R406)63,
Ken Duncan and Peter Warner are at the Bill
paclitaxel64, rapamycin65 and cyclosporine66.
& Melinda Gates Foundation, PO Box 23350,
Conversely, there are numerous examples Seattle, Washington 98102, USA.
in which chemical series were pursued far Takushi Kaneko is at the Global Alliance for TB Drug
too long (in hindsight), soaking up valuable Development (TB Alliance), 40 Wall Street, 24th Floor,
resources, time and careers that should have New York, New York 10005, USA.
been invested elsewhere. As an additional Kiyoshi Kita is at the University of Tokyo, 7‑3‑1,
complication in navigating between these two Hongo, Bunkyo, Tokyo 113–0033, Japan.
extremes, the TPPs may shift with changes in Charles E. Mowbray is at the Drugs for Neglected
the clinical landscape for infectious diseases Diseases initiative (DNDi), 15 Chemin Louis-Dunant,
1202 Geneva, Switzerland.
that affect people in the developing world; for
Correspondence to K.K. 
example, such changes may include a shift in
e-mail: kei.katsuno@ghitfund.org
policy from curing patients ad hoc to eradi-
doi:10.1038/nrd4683
cation programmes that require mass drug Published online 5 October 2015
administration, a decision to focus on the
1. World Health Organization. Antimicrobial resistance:
prevention of transmission rather than global report on surveillance 2014 (WHO, 2014).
on cures, changing disease priorities and 2. Martis, E. A., Radhakrishnan, R. & Badve, R. R.
High-throughput screening: the hits and leads of drug
pharmaco­economics, the emergence of discovery — an overview. J. Appl. Pharma. Sci. 1,
co‑infections (with concomitant drug–drug 2–10 (2011).
3. Khanna, I. Drug discovery in pharmaceutical industry:
interaction risks) or the spread of resistance. productivity challenges and trends. Drug Discov.
Nevertheless, in the midst of all these com- Today 17, 1088–1102 (2012).
4. Nwaka, S. & Hudson, A. Innovative lead discovery
plexities, project teams can attempt to steer strategies for tropical diseases. Nat. Rev. Drug Discov.
their compounds using the set of criteria and 5, 941–955 (2006).
© 2015 Macmillan Publishers Limited. All rights reserved
PERSPECTIVES
5. Nwaka, S. et al. Advancing drug innovation for
neglected diseases-criteria for lead progression.
PLoS Negl Trop. Dis. 3, e440 (2009).
6. Burrows, J. N., Hooft van Huijsduijnen, R., Möhrle, J. J.,
Oeuvray, C. & Wells, T. N. C. Designing the next
generation of medicines for malaria control and
eradication. Malar. J. 12, 187 (2013).
7. Swinney, D. C. Phenotypic versus target-based drug
discovery for first‑in‑class medicines. Clin. Pharmacol.
Ther. 93, 299–301 (2013).
8. Eder, J., Sedrani, R. & Wiesmann, C. The discovery of
first‑in‑class drugs: origins and evolution. Nat. Rev.
Drug Discov. 13, 577–587 (2014).
9. Payne, D. J., Gwynn, M. N., Holmes, D. J. &
Pompliano, D. L. Drugs for bad bugs: confronting
the challenges of antibacterial discovery. Nat. Rev.
Drug Discov. 6, 29–40 (2007).
10. Deng, X. et al. Fluorine modulates species selectivity
in the triazolopyrimidine class of Plasmodium
falciparum dihydroorotate dehydrogenase inhibitors.
J. Med. Chem. 57, 5381–5394 (2014).
11. Schuster, F. L. Cultivation of Plasmodium spp.
Clin. Microbiol. Rev. 15, 355–364 (2002).
12. Mutai, B. K. & Waitumbi, J. N. Apoptosis stalks
Plasmodium falciparum maintained in continuous
culture condition. Malar J. 9 (Suppl. 3), S6 (2010).
13. Chatelain, E. Chagas disease drug discovery:
toward a new era. J. Biomol. Screen 20, 22–35 (2015).
14. Wells, T. N. C., Hooft van Huijsduijnen, R. & Van
Voorhis, W. C. Malaria medicines: a glass half full?
Nat. Rev. Drug Discov. 14, 1–18 (2015).
15. Bosman, A. & Mendis, K. N. A major transition in
malaria treatment: the adoption and deployment
of artemisinin-based combination therapies.
Am. J. Trop. Med. Hyg. 77, 193–197 (2007).
16. Leibert, E., Danckers, M. & Rom, W. N. New drugs to
treat multidrug-resistant tuberculosis: the case for
bedaquiline. Ther. Clin. Risk Manag. 10, 597–602
(2014).
17. Xavier, A. S. & Lakshmanan, M. Delamanid: a new
armor in combating drug-resistant tuberculosis.
J. Pharmacol. Pharmacother. 5, 222–224 (2014).
18. Wunberg, T. et al. Improving the hit‑to‑lead process:
data-driven assessment of drug-like and lead-like
screening hits. Drug Discov. Today 11, 175–180 (2006).
19. Hughes, M. et al. Early drug discovery and development
guidelines: for academic researchers, collaborators, and
start‑up companies. Assay Guidance Manual [online],
http://www.ncbi.nlm.nih.gov/books/NBK92015/ (2012).
20. Ariey, F. et al. A molecular marker of artemisinin-
resistant Plasmodium falciparum malaria. Nature
505, 50–55 (2014).
21. Ashley, E. A. et al. Spread of artemisinin resistance in
Plasmodium falciparum malaria. N. Engl. J. Med.
371, 411–423 (2014).
22. Burrows, J. Microbiology: malaria runs rings round
artemisinin. Nature 520, 628–630 (2015).
23. Mbengue, A. et al. A molecular mechanism of
artemisinin resistance in Plasmodium falciparum
malaria. Nature 520, 683–687 (2015).
24. Burrows, J. N. & Waterson, D. in Third World Diseases
(ed. Elliot, R. L.) 125–180 (Springer, 2011).
25. Gamo, F. J. et al. Thousands of chemical starting
points for antimalarial lead identification. Nature
465, 305–310 (2010).
26. Guiguemde, W. A. et al. Chemical genetics of
Plasmodium falciparum. Nature 465, 311–315 (2010).
27. Plouffe, D. et al. In silico activity profiling reveals the
mechanism of action of antimalarials discovered in a
high-throughput screen. Proc. Natl Acad. Sci. USA
105, 9059–9064 (2008).
28. Meister, S. et al. Imaging of Plasmodium liver stages
to drive next-generation antimalarial drug discovery.
Science 334, 1372–1377 (2011).
29. Angulo-Barturen, I. et al. A murine model of
falciparum-malaria by in vivo selection of competent
strains in non-myelodepleted mice engrafted with
human erythrocytes. PLoS ONE 3, e2252 (2008).
30. Nilsen, A. et al. Quinolone‑3‑diarylethers: a new class
of antimalarial drug. Sci. Transl. Med. 5, 177ra37
(2013).
31. Winzeler, E. A. & Manary, M. J. Drug resistance
genomics of the antimalarial drug artemisinin.
Genome Biol. 15, 544 (2014).
32. Isozumi, R. et al. Novel mutations in K13 propeller
gene of artemisinin-resistant Plasmodium falciparum.
Emerg. Infect. Dis. 21, 490–492 (2015).
33. Tun, K. M. et al. Spread of artemisinin-resistant
Plasmodium falciparum in Myanmar: a cross-sectional
survey of the K13 molecular marker. Lancet Infect.
Dis. 15, 415–421 (2015).
8 | ADVANCE ONLINE PUBLICATION www.nature.com/reviews/drugdisc
34. Dembele, L. et al. Towards an in vitro model of
Plasmodium hypnozoites suitable for drug discovery.
PLoS ONE 6, e18162 (2011).
35. Sinden, R. E., Carter, R., Drakeley, C. & Leroy, D.
The biology of sexual development of Plasmodium:
the design and implementation of transmission-
blocking strategies. Malar J. 11, 70 (2012).
36. World Health Organization. Global tuberculosis
report 2014. WHO [online], http://www.who.int/tb/
Publications/Global_Report/en/ (2014).
37. Franzblau, S. G. et al. Rapid, low-technology MIC
determination with clinical Mycobacterium
tuberculosis isolates by using the microplate Alamar
Blue assay. J. Clin. Microbiol. 36, 362–366 (1998).
38. Cho, S. H. et al. Low-oxygen-recovery assay for
high-throughput screening of compounds against
nonreplicating Mycobacterium tuberculosis.
Antimicrob. Agents Chemother. 51, 1380–1385
(2007).
39. Mak, P. A. et al. A high-throughput screen to identify
inhibitors of ATP homeostasis in non-replicating
Mycobacterium tuberculosis. ACS Chem. Biol. 7,
1190–1197 (2012).
40. Wayne, L. G. In vitro model of hypoxically induced
nonreplicating persistence of Mycobacterium
tuberculosis. Methods Mol. Med. 54, 247–269 (2001).
41. Lakshminarayana, S. B. et al. Comprehensive
physicochemical, pharmacokinetic and activity
profiling of anti‑TB agents. J. Antimicrob. Chemother.
70, 857–867 (2015).
42. Dhar, N. & McKinney, J. D. Microbial phenotypic
heterogeneity and antibiotic tolerance. Curr. Opin.
Microbiol. 10, 30–38 (2007).
43. Mitchison, D. A. The search for new sterilizing anti-
tuberculosis drugs. Front. Biosci. 9, 1059–1072 (2004).
44. Franzblau, S. G. et al. Comprehensive analysis of
methods used for the evaluation of compounds
against Mycobacterium tuberculosis.
Tuberculosis (Edinb.) 92, 453–488 (2012).
45. Silva-Miranda, M. et al. High-content screening
technology combined with a human granuloma model
as a new approach to evaluate the activities of drugs
against Mycobacterium tuberculosis. Antimicrob.
Agents Chemother. 59, 693–697 (2015).
46. Pethe, K. et al. Discovery of Q203, a potent clinical
candidate for the treatment of tuberculosis. Nat. Med.
19, 1157–1160 (2013).
47. Russell, D. G., Cardona, P. J., Kim, M. J., Allain, S. &
Altare, F. Foamy macrophages and the progression of
the human tuberculosis granuloma. Nat. Immunol.
10, 943–948 (2009).
48. Harper, J. et al. Mouse model of necrotic tuberculosis
granulomas develops hypoxic lesions. J. Infect. Dis.
205, 595–602 (2012).
49. Datta, M. et al. Anti-vascular endothelial growth
factor treatment normalizes tuberculosis granuloma
vasculature and improves small molecule delivery.
Proc. Natl Acad. Sci. USA 112, 1827–1832 (2015).
50. Hawn, T. R., Shah, J. A. & Kalman, D. New tricks
for old dogs: countering antibiotic resistance in
tuberculosis with host-directed therapeutics.
Immunol. Rev. 264, 344–362 (2015).
51. Moore, E. M. & Lockwood, D. N. Treatment of visceral
leishmaniasis. J. Glob. Infect. Dis. 2, 151–158 (2010).
52. Sundar, S. & Chakravarty, J. An update on
pharmacotherapy for leishmaniasis. Expert Opin.
Pharmacother. 16, 237–252 (2014).
53. Molina, I. et al. Randomized trial of posaconazole
and benznidazole for chronic Chagas’ disease.
N. Engl. J. Med. 370, 1899–1908 (2014).
54. De Rycker, M. et al. Comparison of a high-throughput
high-content intracellular Leishmania donovani assay
with an axenic amastigote assay. Antimicrob. Agents
Chemother. 57, 2913–2922 (2013).
55. Keenan, M. et al. Selection and optimization of hits
from a high-throughput phenotypic screen against
Trypanosoma cruzi. Future Med. Chem. 5, 1733–1752
(2013).
56. Pena, I. et al. New compound sets identified from
high throughput phenotypic screening against three
kinetoplastid parasites: an open resource. Sci. Rep.
5, 8771 (2015).
57. Don, R. & Ioset, J. R. Screening strategies to identify
new chemical diversity for drug development to treat
kinetoplastid infections. Parasitology 141, 140–146
(2014).
58. Lipinski, C. A., Lombardo, F., Dominy, B. W. &
Feeney, P. J. Experimental and computational
approaches to estimate solubility and permeability
in drug discovery and development settings.
Adv. Drug Deliv. Rev. 46, 3–26 (2001).
© 2015 Macmillan Publishers Limited. All rights reserved
59. Gilbert, I. H. Drug discovery for neglected diseases:
molecular target-based and phenotypic approaches.
J. Med. Chem. 56, 7719–7726 (2013).
60. Ansari, M. T. et al. Malaria and artemisinin derivatives:
an updated review. Mini Rev. Med. Chem. 13,
1879–1902 (2013).
61. Clark, R. L. Embryotoxicity of the artemisinin
antimalarials and potential consequences for use in
women in the first trimester. Reprod. Toxicol. 28,
285–296 (2009).
62. Toovey, S. Safety of artemisinin antimalarials.
Clin. Infect. Dis. 42, 1214–1215 (2006).
63. Braselmann, S. et al. R406, an orally available spleen
tyrosine kinase inhibitor blocks fc receptor signaling
and reduces immune complex-mediated inflammation.
J. Pharmacol. Exp. Ther. 319, 998–1008 (2006).
64. Oberlies, N. H. & Kroll, D. J. Camptothecin and taxol:
historic achievements in natural products research.
J. Nat. Prod. 67, 129–135 (2004).
65. Tsang, C. K., Qi, H., Liu, L. F. & Zheng, X. F. Targeting
mammalian target of rapamycin (mTOR) for health and
diseases. Drug Discov. Today 12, 112–124 (2007).
66. Borel, J. F. History of the discovery of cyclosporin
and of its early pharmacological development.
Wien Klin. Wochenschr. 114, 433–437 (2002).
67. Slingsby, B. T. & Kurokawa, K. The Global Health
Innovative Technology (GHIT) fund: financing medical
innovations for neglected populations. Lancet Glob.
Health 1, e184–185 (2013).
68. Holmes, D. The GHIT fund shows its cards. Nat. Rev.
Drug Discov. 12, 894 (2013).
69. Crump, A. & Omura, S. Ivermectin, ‘wonder drug’ from
Japan: the human use perspective. Proc. Jpn Acad.,
Ser. B 87, 13–28 (2011).
70. Kita, K., Shiomi, K. & Omura, S. Advances in drug
discovery and biochemical studies. Trends Parasitol.
23, 223–229 (2007).
71. Omura, S. & Crump, A. The life and times of ivermectin -
a success story. Nat. Rev. Microbiol. 2, 984–989 (2004).
72. Strader, C. R., Pearce, C. J. & Oberlies, N. H.
Fingolimod (FTY720): a recently approved multiple
sclerosis drug based on a fungal secondary
metabolite. J. Nat. Prod. 74, 900–907 (2011).
73. Endo, A. A historical perspective on the discovery of
statins. Proc. Jpn Acad. Ser. B Phys. Biol. Sci. 86,
484–493 (2010).
74. Tobert, J. A. Lovastatin and beyond: the history of the
HMG-CoA reductase inhibitors. Nat. Rev. Drug Discov.
2, 517–526 (2003).
75. Baell, J. B. & Holloway, G. A. New substructure filters
for removal of pan assay interference compounds
(PAINS) from screening libraries and for their exclusion
in bioassays. J. Med. Chem. 53, 2719–2740 (2010).
76. McGovern, S. L., Caselli, E., Grigorieff, N. &
Shoichet, B. K. A common mechanism underlying
promiscuous inhibitors from virtual and high-throughput
screening. J. Med. Chem. 45, 1712–1722 (2002).
77. Baell, J. & Walters, M. A. Chemistry: chemical con artists
foil drug discovery. Nature 513, 481–483 (2014).
78. Burrows, J. N., Leroy, D., Lotharius, J. & Waterson, D.
Challenges in antimalarial drug discovery. Future Med.
Chem. 3, 1401–1412 (2011).
79. Ruecker, A. et al. A male and female gametocyte
functional viability assay to identify biologically relevant
malaria transmission-blocking drugs. Antimicrob.
Agents Chemother. 58, 7292–7302 (2014).
80. Orme, I. Cellular and genetic mechanisms underlying
susceptibility of animal models to tuberculosis
infection. Novartis Found. Symp. 217, 112–117;
discussion 117–119 (1998).
81. Priest, B. T., Bell, I. M. & Garcia, M. L. Role of hERG
potassium channel assays in drug development.
Channels (Austin) 2, 87–93 (2008).
82. Riss, T. L. et al. Cell viability assays. Assay Guidance
Manual [online], http://www.ncbi.nlm.nih.gov/books/
NBK53196/ (2004).
Competing interests statement
The authors declare no competing interests.
FURTHER INFORMATION
Drugs for Neglected Diseases initiative: http://www.dndi.org
Global Health Innovative Technology (GHIT) Fund:
http://www.ghitfund.org
Japanese Pharmaceutical Manufacturers Association:
http://www.jpma.or.jp
Medicines for Malaria Venture: http://www.mmv.org/
TB Alliance: http://www.tballiance.org/
WHO website: http://www.who.int/tb
ALL LINKS ARE ACTIVE IN THE ONLINE PDF